The relative potency of standardized grass extracts (orchard, redtop, Bermuda, perennial rye, and timothy) with respect to their respective FDA reference extracts were estimated by both in vivo PLST and in vitro RAST. The concordance between PLST and RAST relative potency was found to be 90.6%. For each grass except timothy, the PLST and RAST relative potency estimates fell within their respective 95% confidence interval demonstrating equipotence by each assay. For timothy grass, the PLST relative potency was just below the 95% lower limit while the RAST relative potency was just above the 95% lower limit. Nonetheless, the concordance of relative potency for timothy between PLST and RAST was 92%. There was no statistically significant difference found between the RAST and PLST relative potency estimates for all the grasses tested. These data indicate that RAST relative potency estimates for standardized grass extracts are highly predictive of their in vivo relative potency. The relative potency of standardized short ragweed extracts estimated based on Amb a 1 content with respect to the CBER reference was compared with the relative potency based on parallel-line skin test. The correlation coefficient of the two estimates was 0.995, p<0.001. The skin test estimate of RP was 105% of the Amb a 1 estimate underrating that potency based on Amb a 1 is highly predictive of clinical activity. The relative potency of standardized cat extracts based on Fel d 1 content with respect to the CBER reference was compared with the relative potency in Fel d 1 reactors estimated by parallel line skin test. The coefficient was 0.85, p=0.0002. The mean in vivo relative potency estimate was 91.4$ of the in vitro estimate confirming the high predictive value in estimating clinical potency based on Fel d 1.